This invention relates generally to combined cycle power plants that may or may not incorporate cogeneration into their cycle. As will be demonstrated by the following disclosure, the increasing need for more energy efficient and environmentally friendly methods of generating power has prompted a widespread search for systems and methods to achieve these goals. However, current technologies have a generally myopic view of the total economic impact imposed by a concentration on energy efficiency and environmental issues alone.
The present invention proposes to break with tradition and include as part of the economic and environmental analysis the complete equipment complement required to implement a desired plant load (power) rating. By incorporating this analysis into a new system and method of supplemental firing and heat recovery, the present invention dramatically cuts the overall economic and environmental cost of installed power plants by reducing the equipment complement while maintaining or reducing plant emissions. The result of this improvement over the art is cheaper and cleaner electrical energy than would be possible using conventional combined cycle plants that are currently known in the art.
Overview
Combined cycle power plants and cogeneration facilities utilize gas turbines (GT(s)) as prime movers to generate power. These GT engines operate on the Brayton Cycle thermodynamic principle and typically have high exhaust flows and relatively high exhaust temperatures. These exhaust gases, when directed into a heat recovery boiler (typically referred to as a heat recovery steam generator (HRSG)), produce steam that can be used to generate more power and/or provide process steam requirements. For additional power production the steam can be directed to a steam turbine (ST) that utilizes the steam to produce additional power. In this manner, the GT produces work via the Brayton Cycle, and the ST produces power via the Rankine Cycle. Thus, the name xe2x80x9ccombined cyclexe2x80x9d is derived. In this arrangement, the GT Brayton Cycle is also referred to as the xe2x80x9ctopping cyclexe2x80x9d and the ST Rankine Cycle is referred to as the xe2x80x9cbottoming cycle,xe2x80x9d as the topping cycle produces the energy needed for the bottoming cycle to operate. Thus, the functionality of these cycles is linked in the prior art.
Rankine Cycle
Steam has been used for power applications for more than a century. Early applications utilized a pump to bring the water up to the desired pressure, a boiler to heat the water until it turned to steam, and a steam engine, typically a piston type engine, to produce shaft horsepower. These power plants were used in factories, on locomotives, onboard steamships, and other power applications.
As technology progressed, the trend for the use of steam engines diminished and the use of steam turbines increased. One advantage of the steam turbine was its overall cycle efficiency when used in conjunction with a condenser. This allowed the steam to expand significantly beyond normal atmospheric pressure down to pressures that were only slightly above an absolute vacuum (0.5 to 2 pounds per square inch absolute (psia)). This allowed the steam to expand further than in an atmospheric exhaust configuration, extracting more energy from a given mass of steam, thus producing more power and increasing overall steam cycle efficiency. This overall steam cycle, from a thermodynamic perspective, is referred to as the Rankine Cycle.
FIG. 1 illustrates the thermodynamic operation of the Rankine Cycle. In FIG. 1, graph (100) illustrates the Rankine Cycle on a Pressure versus Volume plot. From point (101) to point (102), water is pressurized at constant volume. From point (102) to point (103), the water is boiled into steam at constant pressure. Point (103) to point (104) defines the process where the steam expands isentropically and produces work. Then, from point (104) to point (101) the low-pressure steam is condensed back to water and the cycle is complete.
Also in FIG. 1, graph (110) illustrates the Rankine Cycle on a Temperature versus Entropy plot. From point (111) to point (112), water is pressurized. From point (112), the water is boiled into steam at constant temperature until it is all steam, then it is superheated to point (113). Point (113) to point (114) defines the process where the steam expands isentropically and produces work. From point (114) to point (111) the low-pressure steam is condensed back to water at constant temperature to complete the cycle. See Eugene A. Avallone and Theodore Baumeister III, MARKS"" STANDARD HANDBOOK FOR MECHANICAL ENGINEERS (NINTH EDITION) (ISBN 0-07-004127-X, 1987) in Section 4-20 for more discussion on the Rankine Cycle.
Power Plant Cycle
For a number of decades, the Rankine Cycle has been used to produce most of the electricity in the United States, as well as in a number of other countries. FIG. 2 illustrates a schematic of the basic Rankine Cycle, with the four primary components being the Boiler Feed Pump (BFP) (201), Boiler evaporator/superheater (BOIL) (203, 205), Steam Turbine (ST) (207), and the Condenser (COND) (209). Note that either one or multiples of any component are possible in the arrangement, but for simplicity, only one of each is shown in FIG. 2. The sub-critical Rankine Cycle (steam pressures less than 3206.2 psia) starts as water at the inlet (211) of the BFP (201). The water is then pumped to a desired discharge pressure by the BFP (201). This pressurized water (202) is then sent to the evaporator (EVAP) (203) where heat is added to the pressurized water. Typically this is accomplished by burning a fuel in the boiler, and the heat of combustion is then transferred to the pressurized water that is routed through tubes and other passages and/or vessels in the boiler. As sufficient heat is added to the pressurized water, it boils and turns into steam (204). This steam now exists in the two-phase region where both steam and water coexist at the same pressure and temperature, called the saturation pressure and saturation temperature. For most applications designed in recent decades, this steam (204) is then sent to a superheater section (SHT) (205) in the boiler where it is heated to a higher temperature than saturation temperature. This steam (206) is now referred to as superheated steam. Superheated steam reduces (but does not eliminate) the risk of water carryover into the steam turbine (207), which is of concern since water carryover can cause extensive internal steam turbine damage. Of more importance, however, is the fact that superheated steam yields better cycle efficiencies. This is of great importance to large central power stations.
Once produced, the superheated steam (206) is sent to the steam turbine (207), typically via one or more pipes. The steam then begins to expand in the steam turbine (ST) and produce shaft horsepower. After traveling through the steam turbine down to a low exhaust pressure, the steam exits the ST (208), and is sent to the condenser (209), where it is then condensed back into water. This device is typically a tubed heat exchanger, but can also be other types of heat exchangers such as a spray chamber, air-cooled condenser, or other heat exchange device used for a similar purpose. After rejecting heat from the low-pressure steam and condensing the steam back to water, the condenser collects the water in an area commonly referred to as the hotwell (HW) (210), where it is then typically pumped through the condensate line (211) and back to the BFP (201). Shaft horsepower produced in the ST is converted into electrical power in the generator (GEN) (212). This cycle of one unit of water from the point of beginning, through the system, and back to the point of origin defines the basic Rankine Cycle.
Current power plants using only steam as the motive fluid typically use a boiler to produce the steam. This boiler may be fueled by a variety of fuels, including oil, natural gas, coal, biomass, as well as others, such as nuclear fuel. The boilers may also use a combination of fuels as well. Depending upon capital cost considerations, fuel costs, maintenance issues, and other factors, the owners and engineers will select the steam pressure and temperature at which the boiler will produce steam.
Due to the size and weight of large steam turbines, they require extended periods for start-up. This is due to the thick metal casings and large heavy rotors that are utilized in their construction. Therefore, these machines require long start-up periods to allow these heavy components to warm up uniformly, and avoid interference between stationary and rotating parts that may occur due to differential thermal expansion.
Although the heavy construction is a deterrent to rapid startup, it provides for robust construction and sustained performance levels. Even after four (4) years of nearly continuous service, the performance decay for a large ST should be less than 2%. This performance decay, combined with the fact that the boiler feed pumps only consume about 2% of the ST output, mean that the performance levels for a ST sustain near optimum levels for extended periods of time, even with decay in the auxiliary loads (BFP). In other words, if the BFP efficiency decays from 75% to 65%, the auxiliary load only increases from 2.00% to 2.31%. This is a small effect on the net output of the Rankine cycle plant, and is another one of its major advantages.
Brayton Cycle
The Brayton Cycle varies quite differently from the Rankine Cycle, as a major part of the cycle involves the compression of the working fluid, which is a compressible gas. This process consumes a great deal of power, therefore, efficient compression of the working fluid is essential to an efficient Brayton Cycle.
Common engines that utilize a Brayton Cycle are aircraft turboprops, jet engines, and gas turbines for stationary application. These engines work by ingesting air (the working fluid), compressing it to a higher pressure, typically 3 to 30 times that of the surrounding ambient air, adding heat through direct combustion (although heat addition from an external source is also possible), and then expanding the resulting high-pressure hot gases through a turbine section. Aircraft engines primarily produce thrust to propel an aircraft through the air. Therefore, some or perhaps none of their output is in the form of shaft horsepower (a turboprop gas turbine engine may drive the propeller, but may also produce some thrust from the high velocity exhaust gases).
For stationary gas turbine applications, the purpose of the engine is to produce shaft horsepower. Approximately ⅔ of the energy produced by the turbine section of the gas turbine is required to drive the compressor section, with the remaining ⅓ available to drive a load. This drawback of GT systems may be used to advantage in the present invention as described later in this document.
Aircraft engines utilize the Brayton Cycle because these engines offer high thrust-to-weight ratios. This is needed to minimize the aircraft weight so it can fly. For stationary applications, gas turbines are used to provide electrical power at peak loads. This is another advantage the Brayton Cycle engines have over Rankine Cycle engines: rapid start and stop times (relatively speaking). Since steam turbines are large heavy engines, it is necessary to start them slowly, and allow the heat to slowly soak into the thick casings so as to avoid thermal distortion and potential rubs between the stationary components and rotating components of the engine. A large power plant steam turbine may require a 24-hour warm-up sequence from cold start to reach full load. However, due to the lower operating pressures and lighter weights, gas turbines can be started and brought to fall load within a matter of minutes of start-up.
Therefore, many utilities in the United States and other countries use gas turbines to provide electrical power during peak demand. These turbines are not very efficient in simple cycle (25% to 30% LHV), but meet the electrical demand requirements for a few hours each day.
Steam Turbine Design
When designing a steam turbine for a power plant application (constant speed), the steam turbine design engineer first examines the output rating desired by the customer. This is because the steam turbine will be custom designed and manufactured for the customer to his specification. The steam turbine will not be totally designed from a clean sheet of paper as may be inferred by xe2x80x9ccustomxe2x80x9d, but will utilize components from a xe2x80x9cfamilyxe2x80x9d of hardware and have a unique steam path for the application. After turbine rating, the ST design engineer will look at the plant steam conditions, and based upon these parameters determine an inlet flow to the turbine high-pressure (HP) section. Utilizing this information, the ST design engineer can select the optimum HP casing for the application. In a similar fashion, he can also select the optimum intermediate pressure (IP) and low-pressure (LP) casings as well.
Knowing which casings to use, the engineer then selects the appropriate blading (both stationary and rotating) for the application. This blading size is determined primarily by the volume flow (as opposed to mass flow) of steam through the turbine. With casings and blading determined, the engineer completes the ST design by selecting valves, controls, instrumentation, and other accessories required for operation of the ST. The final design is a high efficiency ST optimized for the customer""s steam conditions and desired rating.
An interesting note concerning this design philosophy is that two STs with the same steam conditions but with large differences in rating (for example, 200 MW versus 400 MW) may actually appear almost identical when viewed from the outside. This is because the optimum casings selected were designed to cover the flow range of both units. However, due to the large volume flow differences, the large unit would have blades that are approximately twice the size (height) internally. It is interesting to note, however, that both these units might have nearly the same HP and IP casings. This means that the larger ST, even with a dramatic increase in rating, may be only incrementally more expensive to manufacture than the ST with the lower rating. This fact may be used to advantage in the present invention as described later in this document.
Gas Turbine Design
Unlike the steam turbine, the gas turbine is not a custom designed machine for each customer. Although accessories such as the starting means, lube oil cooler type, and control options may be specified by the customer for a particular application, the core engine is essentially standard. Much of this is due to the fact that the gas turbine is actually a packaged power plant, which needs essentially only fuel to produce power. In contrast, the steam turbine is merely a component of a power plant, and requires a boiler, BFP, and condenser to become a complete power plant. Therefore, the gas turbine compressor section, combustion system, and turbine section must all be designed to work together. Since the design of the GT is a highly intensive engineering task, GT designs are generally completed and extensively tested, after which they are mass produced without variation to the core engine design. This eliminates the customer""s ability to specify power output for either a facility with gas turbines only or a combined cycle facility in the prior art. When building a combined cycle plant, the customer simply must choose from a selection of standard offerings by a manufacturer that best meets his needs for power output, efficiency, and cost.
Steam Turbine/Gas Turbine Efficiency and Rating Comparison
The largest and most efficient GT available today for 60-cycle power production is rated at approximately 250 MW with an efficiency of 40.0% LHV (Lower Heating Value). An example of this GT is the Westinghouse model 501G. This is in contrast to STs that can be rated up to as high as 1500 MW and have overall cycle efficiencies in excess of 45% LHV. Therefore, comparing a Rankine Cycle power plant to a Brayton cycle power plant, where each employs the largest and most efficient turbine available, the single ST Rankine cycle is approximately six (6) times larger in rating and 12.5% more efficient than the Brayton Cycle with its best GT. This fact may be used to advantage in the present invention as described later in this document.
Cogeneration/Combined Cycle
One characteristic of the gas turbine is that it expels high volumes of exhaust gases at high temperature. With the advent of the Arab oil embargo of 1973 and higher energy prices, more focus was put on finding ways to utilize the energy contained in these high temperature exhaust gases.
Significantly higher energy prices in the early 1970s signaled the start of a wave of small power plants built using the principles of cogeneration. Cogeneration is defined as the simultaneous production of mechanical or electrical energy in conjunction with thermal energy. In other words, the utilization of an engine (gas turbine or otherwise) to produce power, while at the same time using waste heat from the engine for another process, thus displacing fuel that would otherwise be used for said process. This was a very efficient method from a fuel utilization perspective and was encouraged by the United States Public Utilities Regulation and Policies Act (PURPA) of 1978, which mandated that the local utilities must purchase power from qualified cogenerators, and buy it at a rate which included avoided cost for new power plants.
At first cogeneration projects were small, typically less than 50 MW. They consisted of small gas turbines with a HRSG to produce steam. In many instances, the steam pressures were relatively low (less than 600 psig), as the steam was used for process requirements. Some projects included a steam turbine, while others did not. As the industry matured, larger plants with higher steam pressures were designed to increase bottoming cycle efficiency. In addition, the major gas turbine manufacturers designed and built larger and more efficient gas turbines to meet the needs of the cogeneration marketplace. Soon, due to their high efficiency, low emissions, and low capital cost (dollars per kW of capacity), cogeneration power plants gave way to combined cycle power plants (plants that produced only power and provided no useful thermal energy as was the case with cogeneration plants). Some cogeneration projects are still being proposed and constructed, but they are now typically referred to as combined heat and power (CHP) projects.
Although there was this gradual shift from small cogeneration projects to large combined cycle power plants, the arrangement and overall system and method for producing power was for the most part unchanged. The gas turbine(s) was the primary engine, and a HRSG was utilized to capture the heat in the GT exhaust gases. Optimized for maximum power production, the steam turbine(s) produced additional power equal to approximately 50% of the power produced by the gas turbine(s). The HRSG was typically a two or three pressure level boiler to maximize heat recovery and steam turbine was designed to accept steam from all pressure levels of the HRSG. A review of the manufacturers standard combined cycle offerings will illustrate this trend. The 1997 TURBOMACHINERY HANDBOOK, (USPS 871-500, ISSN 0149-4147), tabulates standard combined cycle power plants available from various manufacturer""s including ABB, General Electric, and Westinghouse. In most every instance, the steam turbine""s output is within the range of 40% to 60% of the gas turbine(s) output. General Electric informative document GER-3567G, 1996, xe2x80x9cGE Heavy-Duty Gas Turbine Performance Characteristics,xe2x80x9d by Frank J. Brooks provides the output for the gas turbines used in their combined cycle power plants.
In summary, the system and method utilized by the major manufacturer""s of combined cycle power plant turbomachinery evolved from the small cogeneration power facilities that were designed to produce both power and thermal energy simultaneously. The sizes for combined cycle power plants have grown from small cogeneration projects under 50 MW to large structured plants producing in excess of 700 MW (as in the Westinghouse 2X1 501G combined cycle). These plants are primarily gas turbine power plants, with the steam turbine producing additional power which is nominally 40% to 60% of the power produced by its associated gas turbine(s). With the gas turbine as the prime engine, the ratings on the standard combined cycle power plants are very rigid, as gas turbines are production line items, versus steam turbines which are largely custom designed and manufactured. A new system and method that offers more flexibility, without compromising the benefits of combined cycle power such as high efficiency, low emissions, and low capital cost, would be welcomed by the industry.
Efficiency Optimizations
Feedwater Heater
With Rankine Cycle plants producing billions of dollars of electricity annually, and consuming commensurate amounts of fuel each year, a great deal of design and analysis has been done to optimize the Rankine Cycle by introducing small variations or revised configurations. FIG. 3 illustrates some of the common variations that are used to design a Rankine Cycle for optimum efficiency. Part (303) of FIG. 3 schematically represents a feedwater heater (FWH). This device is typically a shell and tube heat exchanger, but could be a plate and frame heat exchanger, vortex mixing heat exchanger that mixes the feedwater with small amounts of steam, or other heat exchange device used for a similar purpose. Analysis has proven that utilizing extraction steam from the steam turbine to preheat water before it enters the boiler increases the cycle efficiency.
The feedwater heater (303) uses steam that is extracted from the steam turbine at an optimum point to preheat the water between the condenser (319) outlet and the boiler inlet (306). A second feedwater heater (305) is shown in this example. The number of feedwater heaters and their optimum steam conditions are dependent upon a number of factors including but not limited to steam turbine inlet pressure, steam turbine inlet temperature, reheat steam conditions, feedwater heater effectiveness, and other factors. Typically, the number of feedwater heaters, their design, and the inlet steam conditions for these feedwater heaters must be determined for each power plant due to variations in each power plant""s design and individual conditions.
Reheat
Another variation on the Rankine Cycle used to improve cycle efficiency is the use of reheat. This variation involves expanding steam in the steam turbine from design inlet conditions down to some specified reheat pressure. At this point, some energy has already been extracted from the steam to produce shaft horsepower. This lower energy content steam is then redirected to the boiler where it is reheated to a higher temperature. This higher energy content steam is then sent back to the steam turbine to produce more power. More than one reheat can be utilized in the cycle. Again, for the given design conditions, inlet pressures, inlet temperatures, and other conditions, the reheat is designed for the greatest benefit and increase in cycle efficiency.
Other Factors
Other factors that affect cycle efficiency include inlet steam pressure, inlet steam temperature, and exhaust pressure. Typically, higher inlet pressures and higher inlet temperatures yield higher cycle efficiencies. Lower exhaust pressures typically also yield higher cycle efficiencies. Exhaust pressures are normally limited by ambient factors, such as the temperature of the river water, ambient air, or other fluid used to cool the condenser. This will set the limit for the exhaust pressure, and the condenser and associated equipment will be designed to approach this limit, based upon evaluated parameters such as size, cooling medium available, environmental factors, and cost.
Design Limitations
Inlet pressure and inlet temperature are typically selected by the plant design engineer. However, there are limits that are imposed in these designs. As the inlet pressures are increased, the stresses on the boiler tubes, steam turbine casing, and steam turbine internals are increased. These stresses impose limits on the manufacturer""s ability to produce this equipment, or economic limitations on the feasibility of producing this equipment. In addition, above 3206 psia, steam no longer can coexist as both water and steam. This point is referred to as the critical point of steam, and above this pressure steam does not boil. Instead, both water and steam are a fluid and a more intricate super-critical boiler is required to produce steam above this pressure. At higher temperatures, the allowable stress of the boiler tubes, steam turbine casing, and steam turbine internals is reduced, and near the current limits, conventional steam turbine materials rapidly loose their properties as the temperature is increased only small amounts (50xc2x0 F.). Conventional large steam turbines built as state of the art machines have HP inlet temperature limits in the range of 1050xc2x0 F.
Steam Cycle Optimization
Once a boiler steam pressure and temperature is selected, the steam cycle then must be optimized. A typical high efficiency steam cycle will involve the use of feedwater heaters, a reheater, a reheat steam turbine, boiler feed pumps, and a condenser. A descriptive document on cycle optimization is an informative paper issued by General Electric Company (GE) entitled xe2x80x9cSteam Turbine Cycle Optimization, Evaluation, and Performance Testing Considerationsxe2x80x9d (General Electric Reference GER-3642E, 1996) by James S. Wright. This document provides relative performance variations for different cycle parameters such as pressure, temperature, number of reheats, and number of feedwater heaters.
Rankine Cycle Example
FIG. 3 is a schematic representation of a Rankine Cycle with both feedwater heating and reheat. This sub-critical Rankine Cycle works by providing water to the inlet of the boiler feed pump (BFP) (301). The water is then pumped to a desired discharge pressure by the BFP (301). This pressurized water is then sent through the feedwater line (302) to feedwater heater (FWH) (303) and through line (304) to feedwater heater (305). The feedwater heaters (303, 305) preheat the feedwater before it enters the boiler at the boiler inlet (306). This preheated feedwater travels to the evaporator section (307) of the boiler where heat is added to the pressurized water.
Steam exits the boiler section at (308) and continues to superheater section (309) and exits at (310). This superheated steam is sent to the high-pressure (HP) section of the steam turbine (311). The steam expands through the HP section to (312), and then returns to the reheat section of the boiler (RHT) (313) where heat is added to return the steam typically to a temperature at or near the inlet steam temperature. This reheat steam is then sent to the Intermediate Pressure (IP) section of the steam turbine at (314). This steam then expands through the IP turbine section (315) and produces shaft horsepower. The steam then exits the IP section and via the crossover pipe (316) and goes to the LP section of the steam turbine (317).
Due to the high volume flows at low-pressure, the LP section is typically a double flow section on large units, so steam enters the middle of the casing and travels both forward and aft through the blading to produce more shaft horsepower. The steam then exhausts at (318) into the condenser (COND) (319). Condensed steam leaves the hotwell (330) and returns via the feedwater line (320) to the inlet of the BFP (301). For feedwater heating, steam is extracted from the IP and LP sections of the steam turbine at (321) and (324) and sent to feedwater heaters (305) and (303) respectively via lines (323) and (326). Non-return valves are used in these lines, (322) and (325), to prevent backflow of steam to the ST in case of a trip (emergency shutdown) condition when pressures in the turbine will rapidly drop to condenser pressure. These valves are safety devices only, and are either open or closed. Steam from these extraction lines preheats the feedwater on its way to the boiler. The steam from the extraction lines is condensed in the feedwater heaters and the condensate (327, 328) is returned to the inlet of the BFP (301). Again, shaft horsepower produced in the ST is converted into electrical power in the generator (GEN) (329).
For larger, central power plant applications, typical inlet pressures for sub-critical applications are 1800 and 2400 pounds per square inch gauge (psig). For supercritical applications, pressures of 3500 psig and greater are employed. Inlet steam temperatures for most large steam turbines are limited to about 1050xc2x0 F. for both the inlet and reheat steam. However, some advanced technology steam turbines are utilizing inlet temperatures of 1070xc2x0 F. for the HP inlet and 1112xc2x0 F. for reheat, as detailed in a descriptive document on steam turbines issued by General Electric Company (GE) entitled xe2x80x9cSteam Turbines for Ultrasupercritical Power Plantsxe2x80x9d by Klaus M. Retzlaff and W. Anthony Ruegger (General Electric Reference GER-3945, 1996).
Rankine Cycle Efficiency Comparison
Based upon a steam turbine with a 90% efficiency, FIG. 4 illustrates a relative comparison of a basic Rankine Cycle (Option 1), (excluding boiler efficiency and parasitic power requirements) to one that uses only reheat (Option 2, Option 3), and to one that uses both reheat and feedwater heating (Option 4, Option 5). Variations in the inlet pressure with reheat (Option 3) and feedwater heating (Option 5) are also included. Option 6 and Option 7 are for supercritical steam applications. Option 6 is a supercritical steam cycle with ultrasupercritical (inlet or reheat temperatures above 1050xc2x0 F.) steam conditions and double reheat (steam is reheated twice, at two separate pressure levels, in the boiler). Option 7 is the same as Option 6 with the addition of feedwater heating. For the purposes of this comparison, only two extractions were utilized and the extraction pressures were assumed to be at the cold reheat pressure and the crossover pressure (2nd cold reheat for supercritical applications). More feedwater heaters will yield even better cycle efficiencies. General Electric Company (GE) informative document entitled xe2x80x9cSteam Turbine Cycle Optimization, Evaluation, and Performance Testing Considerationsxe2x80x9d (General Electric Reference GER-3642E, 1996) by James S. Wright provides data for the selection of the optimum number of feedwater heaters, stating that a 1.5% heat rate penalty is assessed for only three feedwater heaters versus seven. Therefore, the feedwater heating cycle efficiency shown on FIG. 4 (Options 4, 5, and 7) has room for improvement. With reheat, optimum feedwater heating, and ultrasupercritical steam conditions, overall plant cycle efficiencies in excess of 45% are possible.
The overall plant cycle efficiency includes not only the basic steam cycle efficiency as shown in FIG. 4, but also the boiler efficiency and parasitic power requirements such as the boiler feed pumps and the condenser circulating water pumps. As stated in POWER MAGAZINE, (ISSN 0032-5929, July/August 1998, page 26):
xe2x80x9cOver the last few years, new designs have evolved to boost efficiencies of steam power plants, and the steam turbine is a large part of this effort. Efficiencies of 45% (LHV) [Lower Heating Value] or higher are now possible with the latest fossil-fired steam plants using the highest steam parameters, advanced feedwater heating cycles, boiler and turbine metallurgies, etc.xe2x80x9d
To obtain an overall plant efficiency of 45% LHV, including the boiler efficiency and parasitic power requirements, typically means that the basic steam cycle efficiency must be even higher than 45%. With a boiler efficiency of 85%, parasitic power requirements of 2.5%, a ratio of HHV (higher heating value) to LHV (lower heating value) of fuel of 1.11 (typical for natural gas), and a plant efficiency of 45% (LHV), the basic steam cycle efficiency would calculate to
48.9%=0.45/(0.85xc3x97(1xe2x88x920.025)xc3x971.11)xe2x80x83xe2x80x83(1)
As seen from FIG. 4, the use of a reheat steam cycle can increase the basic Rankine Cycle efficiency by 4.79% at the tabulated steam pressures. However, the use of reheat as well as increased inlet pressures and feedwater heating can boost efficiency by at least 10.3% for sub-critical steam conditions. (Note that efficiency improvement is the ratio of a particular option efficiency to the base efficiency. Thus, a 40% efficient cycle would convert 40% of the input energy to electricity. That is twice as much as a 20% efficient cycle. Therefore, the efficiency improvement from a 20% efficient cycle to a 40% efficient cycle is 100%, or twice as much output).
Fuel efficiency is of the utmost importance at power plants and a large central coal-fired power plant may expend approximately US$140 million annually for fuel, assuming a plant rating of 1000 MW, 45% thermal efficiency LHV (lower heating value of the fuel), US$2.00 per million BTU for fuel, and 8500 operating hours per year. Given these facts, even a 1% increase in efficiency will equate to large cost savings in fuel (US$1.4 million annually).
Combined Cycle Application
Although the Rankine Cycle has been well proven, today""s more strict energy and environmental standards require more emphasis be placed on fuel efficiency and low emissions from power plants. As a result, new combined cycle plants are being designed and built.
FIG. 5 is a conceptual schematic for a combined cycle application. In the general sense, combined cycle is not limited to a Brayton Cycle topping cycle and a Rankine Cycle bottoming cycle, but can be any combination of cycles. The topping and bottoming cycles could be the same cycle using different fluids. Either way, FIG. 5 would be applicable. In FIG. 5, the topping cycle fluid (TCF) (501) enters the topping cycle engine (TCE) (502) where fuel (CFT) (503) is added to raise its temperature. The fluid performs work that is converted by the topping cycle engine into shaft horsepower. This shaft horsepower drives the topping cycle load (TCL) (504). This load could be an electrical generator, pump, compressor, or other device that requires shaft horsepower. The exhausted fluid from the topping cycle engine is directed through an exhaust line (505) to a heat recovery device (HRD) (506), and then exhausts to an open reservoir (507).
For this example, the topping cycle is an open cycle. In other words, the topping cycle fluid is taken from a large reservoir and discharges to that same reservoir. The heat recovery device (506) captures a portion of the topping cycle exhaust energy and transfers it to the bottoming cycle fluid (BCF) (508). In this example, the bottoming cycle fluid is heated at three separate pressure levels: a high-pressure line (509), intermediate pressure line (510), and low-pressure line (511). These fluids then travel to the bottoming cycle engine (BCE) (512) where it produces shaft horsepower to drive the bottoming cycle load (BCL) (513). Again, this load could be an electrical generator, pump, compressor, or other device that requires shaft horsepower.
From the bottoming cycle engine, the bottoming cycle fluid enters a heat exchanger (HEX) (514) where heat is rejected. The bottoming cycle fluid then enters a pump or compressor or other fluid transfer device (FTD) (515) where it is then returned to the heat recovery device (506). For this example, the bottoming cycle is a closed cycle, meaning that the bottoming cycle fluid is continuously circulated within a closed loop. There could be more than two cycles in this process, and any of the cycles could be either open or closed loop. This describes the basic fundamentals of a combined cycle application.
HRSG in Combined Cycles
In many cogeneration and combination GT/ST power plants built today, combined cycle plants have come to mean power plants that utilize a Brayton Cycle as the topping cycle and a Rankine Cycle as the bottoming cycle. These plants utilize a gas or combustion turbine (GT) as the prime mover (Brayton Cycle machine), with a boiler at the exhaust of the gas turbine to recover the waste heat. This boiler is typically referred to as either a waste heat boiler (WHB) or a heat recovery steam generator (HRSG). It may also have burners in place to increase the exhaust gas temperature and produce more steam than that available from just the waste heat (supplemental firing). The HRSG produces steam that is then sent to the steam turbine (ST) to produce more power. Due to the high temperatures of the working fluid in the GT (approximately 2400xc2x0 F. for GE industry standard xe2x80x9cFxe2x80x9d-class technology machines and 2600xc2x0 F. for Westinghouse industry standard xe2x80x9cGxe2x80x9d-class technology machines), and recovery of the waste heat, the combined cycle plants are much more fuel efficient than the conventional steam plants. In addition, with advances in GT technology and the use of either distillate oil or natural gas fuel, the emissions from the combined cycle plants are extremely low. FIG. 6 illustrates a typical combined cycle application.
The HRSG is distinctly different from a conventional Rankine Cycle boiler. A Rankine Cycle boiler is fueled by a variety of fuels, including oil, natural gas, coal, biomass, as well as others. These Rankine Cycle boilers may also use a combination of fuels as well. The HRSG may not utilize any fuels at all, but only capture and utilize the exhaust heat from the GT. If it is supplementary fired, the HRSG will require more refined fuels such as natural gas or distillate oil. Solid fuels such as coal and biomass are not typically utilized in these types of boilers.
As seen from FIG. 6, there are numerous sections to the HRSG, including three evaporator sections (one for each pressure level), economizers, superheaters, and a reheater. Sections (601) and (602) are economizers. These are large tubed sections in the HRSG that preheat water before it is converted into steam in the Evaporator. Sections (603), (606), and (609) are LP, IP, and HP evaporators respectively. Sections (604), (605), and (607) are feedwater heaters. Section (608) is the IP superheater while sections (610) and (612) are HP superheaters. Section (611) is the reheater section. These HRSGs are typically very large and heavy pieces of equipment with literally miles of tubes inside.
Steam from each pressure level is utilized in the power plant where required, but essentially, most steam is generated for the purpose of producing additional power in the ST. This means that the lower pressure levels of steam must be introduced or admitted to the ST at the proper point on the ST other than the HP inlet. It also means that the ST must have provisions (openings, nozzles, connections, trip valves, etc.) where this steam may be admitted, and that at the operating conditions the steam pressure in the ST at these connections must be less than the pressure of the steam from the HRSG corresponding boiler sections. Otherwise, steam will not flow into the ST.
As noticed from a comparison of FIG. 6 with FIG. 3, the conventional Rankine Cycle utilizes feedwater heaters that take steam from the ST to preheat feedwater, while the HRSG utilizes the GT exhaust heat to provide this function. Therefore, conventional steam fed feedwater heaters are not typically employed in combined cycle applications. In GE informative document GER-3582E (1996), entitled xe2x80x9cSteam Turbines for STAG(trademark) Combined Cycle Power Systemsxe2x80x9d, M. Boss confirms that feedwater heaters are not utilized in the prior art:
xe2x80x9cExhaust sizing considerations are critical for any steam turbine, but particularly so for combined-cycle applications. There are usually no extractions from the steam turbine, since feedwater heating is generally accomplished within the HRSGxe2x80x9d.
Another modification typically used for combined cycle applications is the use of two boiler feed pumps (630), and (631), typically referred to as the LP and HP BFPs respectively. This arrangement allows the LP pump to provide pressurized water for the LP and IP pressure levels and the HP pump provides water for the HP pressure level, which saves pump horsepower. For large combined cycle applications, the steam turbine/condenser arrangement is similar to the Rankine Cycle depicted in FIG. 3, (although internally, the steam path designs are totally dissimilar).
HRSG/Combined Cycle Disadvantages
General Disadvantages
With current technology, maximum inlet pressures to the steam turbine for combined cycle applications are nominally 1800 psia with inlet steam temperatures near the limit of 1050xc2x0 F. for both the inlet and reheat steam. Some of the disadvantages of this HRSG arrangement for combined cycle applications are as follows:
1. Steam cycle efficiencies are much lower than those of conventional steam power plants.
2. Multiple evaporator sections are required to maximize heat recovery. This results in increased equipment and maintenance costs.
3. Multiple evaporator sections require the plant operators and control systems to monitor and control all boiler (evaporator) drum levels.
4. The HRSGs with the multiple sections are very large, require large amounts of infrastructure building volume, large amounts of floor space, and large foundations to support the weight of the HRSG.
5. The HRSGs are expensive (approximately $10 million for a HRSG that recovers exhaust gas heat from one GE Frame 7 GT).
6. Maintenance increases with the number of components, evaporator sections, controls, and other devices.
7. Low-pressure steam (steam other than the highest pressure steam) has much less ability to produce power in the ST than higher pressure steam.
8. Partial load, off design operation, and other conditions besides the design conditions typically have reduced heat recovery and lower cycle efficiencies.
9. Increased amounts of tubing in the HRSG to enhance heat recovery add flow restriction to the exhaust gases from the GT and this increased back pressure decreases GT output and efficiency.
10. Gas turbine exhaust temperatures are not sufficient to produce some of the elevated steam conditions now used in advanced steam cycles (600xc2x0 C. which is equivalent to 1112xc2x0 F.).
11. Balancing problems in the reheat lines with multiple GTs (typically three or more) make it difficult to utilize large STs in combined cycle power plants in the prior art. For modern, large, and efficient combined cycle plants such as a GE S207FA, the steam turbine rating is approximately 190 MW, which is much smaller than GE""s large steam turbines which can exceed 1200 MW. For more information on large steam turbines, reference the informative paper issued by General Electric Company (GE) entitled xe2x80x9cSteam Turbines for Large Power Applicationsxe2x80x9d by John K. Reinker and Paul B. Mason (General Electric Reference GER-3646D, 1996).
Part Load Operation Inefficiencies
Another disadvantage of the combined cycle application is partial load (part load) operation. As the system to which a power plant is connected reduces its load requirement, the power plant must respond by providing less output. This load modulation allows for a constant speed on the machinery and a constant frequency of power (e.g., 60 Hz in the United States and 50 Hz in Europe). To modulate the load at a combined cycle plant, less fuel is burned in the GT, and the power output is reduced. This typically requires a reduction in the GT firing temperature and/or a reduction in GT airflow.
Part load operation reduces the efficiency of the GT, thus reducing the efficiency of the entire combined cycle plant. FIG. 7 illustrates a typical curve for a large modern GT with inlet guide vanes (IGVs) to modulate inlet airflow. Even with the enhanced part load efficiency gained by the use of IGVs, at 60% load (Generator Outputxe2x80x94Percent Design), the GT consumes over 70% of the fuel required at full load (Heat Consumptionxe2x80x94Percent Design). This represents a 17.5% increase in heat rate (specific fuel consumption). For GTs without IGVs, this decay in performance would be even more pronounced.
To help offset this part load decay, plus provide more power output for a given amount of hardware (sometimes referred to as power density), manufacturers can provide combined cycle power plants with two GTs, each with its own HRSG, feeding into one ST (referred to as a 2-on-1 arrangement). With an arrangement such as this, when the power plant load decreases to slightly less than 50% for a 2-on-1 arrangement (2-GTs, 1-ST), one GT can be shut down, and the remaining GT can return to near 100% output. This mode of operation increases part load efficiency below 50% of total plant load as illustrated graphically in FIG. 8. This graphically illustrates a typical two GT comparison taken from GE informative document GER-3574F (1996), entitled xe2x80x9cGE Combined-Cycle Product Line and Performancexe2x80x9d by David L. Chase, Leroy O. Tomlinson, Thomas L. Davidson, Raub W. Smith, and Chris E. Maslak for a curve of GE combined cycle part load performance with a 2-on-1 arrangement. For a 3-on-1 arrangement, switchover from three to two GTs could occur slightly below 67% load. This still provides for substantial increase in plant heat rate at part load conditions. Note that providing this increase in part load efficiency occurs as a result of higher equipment costs. The prior art has yet to solve the efficiency problem without the addition of more equipment that increases the overall power plant costs.
Supplementary Firing of HRSG
Another solution to add flexibility to the operation of a combined cycle power plant is the use of supplementary firing in the HRSG. This mode of operation is when fuel is burned in the HRSG just after the GT (or at some intermediate point within the HRSG). This increases the temperature of the exhaust gas to the HRSG and produces more steam that can be sent to the ST. This allows the plant to produce more power. However, the plant heat rate increases, and fuel efficiency decreases accordingly. This result is stated by Moore of GE in U.S. Pat. No. 5,649,416. This patent, as well as U.S. Pat. No. 5,428,950 by Tomlinson, is referenced by Rice in U.S. Pat. No. 5,628,183. Therefore, supplementary firing of the HRSG is considered by the manufacturers to be a means to obtain more output, but with a penalty on efficiency. GE informative document GER-3574F (1996) entitled xe2x80x9cGE Combined-Cycle Product Line and Performancexe2x80x9d by David L. Chase, Leroy O. Tomlinson, Thomas L. Davidson, Raub W. Smith, and Chris E. Maslak states that
xe2x80x9cincremental efficiency for power produced by supplemental firing is in the 34-36% range based upon lower heating value (LHV) of the fuel.xe2x80x9d
Also in this GE document, Table 14 indicates that HRSG supplemental firing can increase combined cycle plant output in the prior art by 28%, but only with an increase in overall combined cycle heat rate (specific fuel consumption) of 9%. No technique has been shown in the prior art to eliminate this heat rate penalty associated with supplemental firing.
Additionally, supplemental firing in the prior art can be utilized to achieve higher ST/GT ratios than is typical for conventional combined cycles. However, operation at these high levels of ST/GT output are typically short in duration to meet peak power demands, and long term operation at these ratios is not economical. Therefore, conventional combined cycle power plants that are designed with ST/GT ratios approaching unity do not operate predominantly as Rankine cycle power plants, but do so only to satisfy temporary peak plant loads, and do so with a significant efficiency penalty at all operating conditions.
Gas Turbine Performance Decay
As mentioned in the discussion on the Brayton cycle, approximately ⅔ of the energy produced by the turbine section of the gas turbine is required to drive the compressor section, with the remaining ⅓ available to drive a load. This power consumed by the compressor at 67% of the turbine output, is much higher than the Rankine cycle example where the boiler feed pumps (BFP) only consumed 2% of the turbine power. Therefore, the GT is susceptible to performance decay if the compressor does not maintain optimum efficiency.
For example, a typical efficiency for an axial flow air compressor used with a large GT might be 90%. Therefore, if the compressor requires 67% of the turbine section output, the ideal power (100% efficient) would only be (0.67*0.90)=0.603 or 60.3%. If the compressor efficiency were to decay by 2.5%, its new efficiency would be (0.90*0.975)=0.8775 or 87.75%. The compressor power required would now be (0.603/0.8775)=0.6872 or 68.72%. Turbine net output would be reduced from 33% (1.00-0.67) to 0.3128% (1.00-0.6872). This represents a 5.2% loss in output (0.3128/0.33=0.9479). Therefore, it can be readily seen that small decreases in efficiency for the GT compressor lead to large decreases in efficiency and output for a GT.
The efficiency and rating loss of 5% from the above example is typical of many GTs after about one or two years of operation. This efficiency decay is largely a result of worn clearances in the compressor and erosion of the compressor blade tips. New blades and seals will typically restore the compressor efficiency to almost xe2x80x9cnewxe2x80x9d condition efficiency. However, this is a costly and time consuming repair, and would probably only be done at major inspections, which are scheduled approximately every four years for modern GTs. Therefore, plant owners and operators will need to plan on this performance decay between major overhauls of the GTs.
Candidates for Improvement in the Prior Art
From the foregoing discussion it can be seen that parameters of the current and defined technology that are candidates for improvement may be described as follows:
Flexibility
Due to the electrical load demand in a particular region or marketplace, the electric utility (which distributes electrical power to the end users) determines the need for power based on current demand and future projections. For example, if this load was determined to be 850 MW, in a conventional Rankine cycle configuration the utility/Power Developer would contract with an Architect/Engineering (AE) firm to design and build such a plant. The boiler, pumps, condenser, steam turbine, and all the other plant auxiliaries would then be designed for the specified output of 850 MW. This can be accomplished largely due to the fact that steam turbines are custom designed and manufactured. However, with gas turbines being production line items, and combined cycles being primarily gas turbine based power plants, to achieve the highest efficiencies and best capital cost, a utility and/or power developer can no longer specify just their plant output, but must find the best fit for their needs from the available combined cycle offerings from the various manufacturers. For example, a review of the available combined cycle plants from the 1997 TURBOMACHINERY HANDBOOK, (USPS 871-500, ISSN 0149-4147), indicates that there are no 850 MW combined cycle plants available for 60 HZ applications. Thus, a plant developer""s design flexibility is constrained by the current state of the art of combined cycle power plant equipment. This implies that in certain circumstances the equipment complement for a given power plant installation will not be optimal because of constraints placed on plant equipment configurations by the current state of the art.
Efficiency
Combined cycle power plants are extremely energy efficient compared to other conventional means of producing electricity. However, a large central combined cycle power plant rated for 1000 MW at 55% thermal efficiency LHV (lower heating value of the fuel) operating 8500 hours per year at full load with a fuel cost of US$3.00 per million BTU of fuel will expend approximately US$175 million annually for fuel. Even a 1% increase in efficiency will equate to large savings in fuel (US$1.75 million annually).
In U.S. Pat. No. 4,333,310 issued to Robert Uram, a control method is utilized which monitors the steam temperature to the ST and modulates the afterburner (supplemental firing) to control the temperature of the superheated steam. While providing optimum ST inlet temperatures, this function does little to affect load. In this patent, Uram states
xe2x80x9cIt is desired that the steam turbine be operated in what is called a xe2x80x9cturbine followingxe2x80x9d mode wherein the plant is supplying electrical power to a load, such that the steam turbine follows the gas turbines and each afterburner positively follows a respective gas turbine. In other words, the heat contributed by the afterburner follows the temperature of the gas turbine exhaust gas, and the steam produced by the gases exhausted from the afterburners is used in total by the steam turbine.xe2x80x9d
These teachings of the prior art are in direct contrast to that of the present invention in which the heat contribution via supplemental firing is independent of the gas turbines, and the gas turbines are designed to operate substantially at their optimal full rated capacity.
Installed Cost
Next to fuel costs, the largest cost for a combined cycle plant is typically debt service. Manufacturers, engineering firms, and owners are always interested in finding ways to reduce the installed cost of power plants. At 8% interest and US$450 per kW of capacity, a 1000 MW combined cycle power plant would have a debt service of approximately US$45 million per annum for 20 years. Reducing the capacity cost, in US$/kW, directly reduces the debt service.
Temporary Capacity Extension During Peak Demand Loading
One dilemma that faces power plant owners and utilities is the proper selection of power plant capacity. Selecting a plant that is too small results in power shortages, brownouts, and/or the need to purchase expensive power from other producers. Selecting a plant that is too large results in operation at lower efficiency during part load and increased capital cost per kWh produced. In many situations the problem faced by power plant developers is the need to provide for peak power needs and temporary demand loading. This peak may occur only in certain seasons for a limited span of time. Typically in the summer months during peak hours on the hottest days is the most challenging time for power producers to meet the system load. Having the ability to provide excess capacity during this time period is highly desirable, and in the emerging arena of electrical power deregulation, it may prove to be very lucrative. For example, in the early summer of 1999, power shortages in the Northeast United States have caused concern for the system""s ability to meet peak power demands. Some local newscasts have reported costs for capacity at $30/MWh during normal periods and as high as $500/MWh during peak. However, even much greater capacity costs have been incurred, as reported in POWER MAGAZINE, (ISSN 0032-5929, March/April 1999, page 14): xe2x80x9cReserve margins are down nationwide from 27% in 1992 to 12% in 1998, according to Edison Electric Institute, Washington, D.C., because deregulation uncertainty has caused capacity additions to stall. Last summer""s Midwest [United States] price spikes, up to $7000/MWh, garnered most of the press coverage, but spikes of $6000/MWh also occurred in Alberta . . . xe2x80x9d
However, providing peak power will not be lucrative if the power plant owners have to pay for this capacity, pay the debt service, and yet make revenue on this extra capacity only during a few days of the year. Therefore, power plants that can provide more output than normal during peak demand hours are needed to help supply system load during these peak demands.
Reference FIG. 31B for a graphic illustrating the relative percentage of time that a typical power plant spends in peak, intermediate, and base loading conditions. From this graphic it can be surmised that it would never be profitable to design a power plant to peak loading conditions, as they occur less than 10% of the time. Since prior art power plants are generally incapable of wide variations in peak power output, the only practical option available for present power providers is to purchase power over the electrical grid during times of peak power demand. The present invention teaches a system and method which permits this peak demand to be satisfied without the need for purchasing external power over the electrical grid, thus providing an economic advantage over the prior art.
Non-Local Power Generation/Distribution Reliability Issues
One significant problem with the prior art is that the plant capacity is in general a relatively fixed and narrow range of power generation operation. When peak power demands are placed on the electrical grid, electrical power must be purchased from elsewhere on the grid where electrical demand relative to remote plant capacity is lower. There are several major problems with this mode of providing for peak power by rerouting remotely generated power plant capacity.
First, there exist losses associated with transmission of power from remote sites to the place where the electrical power is being demanded. For example, a hot summer day in New York City may require diversion of power from Canada or the western United States, resulting in significant line losses during transmission.
Second, there is a reliability drawback in purchasing power from distant parts of the grid during periods of peak load. While it is possible to redistribute power, the tradeoff is instability in the electrical grid. What can happen is that small failures in remote parts of the grid can cascade throughout the grid to either cause additional equipment failures or cause instability in the grid voltage. Thus, while purchasing power from remote power plants may alleviate some local reliability problems with respect to providing electric power, the tradeoff is an overall reduction in the reliability of the entire electrical grid. Thus, relatively insignificant events in remote parts of the country can cascade throughout the electrical grid and result in serious electrical failures in major metropolitan areas.
Thus, given the above reliability concerns, it is in general always better to be able to provide electrical power local to the demand for that power. While the existing prior art relies heavily on power sharing and distribution, the present invention opts for the more reliable method of generating the power locally to provide a power generation system that is more efficient and reliable that the current state of the art. It is significant to note that the prior art limitations on plant output during peak load generally preclude local generation of the required peak power demand. This forces traditional power plants to purchase power from remote power plants at a substantial (10xc3x97 to 250xc3x97) price penalty.
Operation and Maintenance Costs
Costs for personnel, fuel, maintenance, water, chemicals, spare parts, and other consumables, including other costs such as taxes and insurance, all contribute to Operation and Maintenance (OandM) costs. As the plant size grows, the amount of equipment increases, and as the complexity of the equipment increases, OandM costs also increase. In the quest for higher efficiency, more elaborate and expensive technology is being utilized in the gas turbines. The maintenance costs associated with exotic new materials, intricate blades, and complex hardware is projected to be significantly more expensive than the slightly less efficient, proven gas turbine hardware and associated plant designs.
To be prepared for an equipment failure, plant owners must retain large quantities of spares on hand at their facility. This constitutes inventory that has high costs in terms of both unused capital and taxes. Methods to reduce OandM costs are always desired by the plant owners and operators.
Fuel Gas Compression
Current projections are that natural gas will have a stable supply and price structure until the year 2010. This fuel is clean, efficient, and inexpensive, and thus is the preferred fuel for combined cycle applications. However, if the power plants are not located in close proximity to major natural gas pipelines, the lower pressure natural gas may have to be compressed to a sufficient pressure to be used in the GT. In addition, the higher efficiency GTs such as the Westinghouse model 501G require higher fuel gas pressure than GTs with lower pressure ratios, such as a GE model PG7241FA GT. This need for higher pressure natural gas requires expensive natural gas compressors that are critical service items (the plant cannot operate without them). These natural gas compressors require frequent maintenance and also consume parasitic power (the power to run the compressors reduces the net power available from the power plant to the grid). Reducing the need for these components reduces the plant installed cost, reduces real estate requirements, improves reliability, and increases the plant net output.
Plant Reliability
Electrical power reliability has become a facet that is demanded by both the residential consumer and industrial user of electricity. Therefore, the technology to produce power must be proven and reliable. In U.S. Pat. No. 5,628,183, Rice proposes a higher efficiency combined cycle power plant. However, this system requires the use of diverters in the HRSG, natural gas reformers, and the use of steam superheated to 1400xc2x0 F. These systems will all add greatly to the installed cost and OandM costs. In addition, to date, boiler tubes, HRSGs and STs have not demonstrated long term reliable operation at elevated temperatures above 1150xc2x0 F., and HRSGs with diverters and natural gas reformers are as yet unproven in the marketplace.
Air Consumption
GT engines consume large quantities of air. A typical combined cycle installation will consume approximately 20 lbs. of air per kW of electricity produced. This equates to approximately 260 cubic feet (at sea level) per kW. This air must be filtered before it enters the GT to prevent foreign object damage in the GT. Periodically, the air filters must be cleaned and/or replaced. This adds to the OandM costs and increases plant downtime (time when the plant is out of service and unavailable to produce power).
In addition, the air consumed by the GT is discharged to the HRSG and then exhausted to atmosphere. As more air is consumed, more air must be exhausted. This represents an efficiency loss as the HRSG exhaust temperature is typically about 180xc2x0 F. In addition, this airflow serves to heat the atmosphere and contribute to local air quality problems.
Plant Emissions
In order to obtain a permit to operate, a power plant must first obtain an air permit. This permit typically states the allowable levels of certain criteria pollutants that a plant may emit. Combined cycle power plants are very clean producers of power compared to other conventional methods, but are typically plagued by one criteria pollutant, nitrous oxides (NOX). This criteria pollutant is usually controlled by steam and/or water injection into the GT, dry low NOX combustion systems, and/or exhaust gas aftertreatment. The exhaust gas aftertreatment typically employed is xe2x80x9cSelective Catalytic Reductionxe2x80x9d (SCR) which essentially works by injecting ammonia (NH.sub.3) into the exhaust gas stream in the presence of a catalyst at a specified temperature range to return the NOX formed by the combustion process into N2 and H2O.
In U.S. Pat. No. 3,879,616 by Baker, et al., U.S. Pat. No. 4,578,944 by Martens et al., and U.S. Pat. No. 5,269,130 by Finckh, et al., the plant load is controlled by changes in the GT output. However, at partial load, GT NOX emissions are typically increased. Therefore, it may be necessary to introduce more ammonia into the exhaust gases for emission reduction. This increases OandM costs, and can be significant to the point where, at the plant design stage, the desired GTs cannot be used due to high emission levels at part load operation. Also, if run at full load, some plants may not require SCR, but due to part load operation, SCR will be required. Another factor related to emissions is air consumption. GTs require large amounts of air, and the more air that is consumed, the more potential there is for emissions.
Environmental Considerations
Besides air emissions, a power plant must be concerned with other environmental impacts as well. To operate a steam plant, a clean source of water must be available to provide make-up water. This make-up water is used to replace steam/water that is lost to ambient through leaks, blowdown, or other loss. Blowdown is the water that is taken from the evaporator sections of the HRSG and dumped to the sewer. This blowdown typically is taken from a low point on the HRSG to remove feedwater that has high concentrations of minerals and deposits. This process helps keep the steam path clean and minimizes ST deposits and blade failure due to stress corrosion cracking. This blowdown must be discharged into rivers, streams, etc. and as such requires water permits that may be difficult and time consuming to obtain from regulatory authorities.
Distributed Plant Control System (DCS)
Modern combined cycle plants typically use a distributed control system (DCS) to control the entire plant. These DCS controls integrate with the individual control systems on the GTs and STs. Many other parameters can be monitored and controlled by the DCS. Use of controls to better either efficiency or operation is described in U.S. Pat. No. 3,879,616 by Baker et al., U.S. Pat. No. 4,201,924 by Uram, and U.S. Pat. No. 4,578,944 by Martens et al. None of these patents, however, provide control of heat transfer in the HRSG. In U.S. Pat. No. 5,269,130 by Finckh et al., a method of controlling excess heat in the HRSG is utilized for part load operation of the GT. This method, however, does not provide comprehensive control, but only a means for recovering low temperature waste heat. None of the aforementioned patents has devised a method to control the exhaust gas temperature of the HRSG to its optimum temperature.
Plant Operational Efficiency
Combined cycle power plants in the prior art that are designed for maximum efficiency typically utilize multi-pressure HRSGs, commonly at three pressure levels. For each HRSG, and for each pressure level, the operations staff must monitor the steam drum level. Also, parameters such as water quality and chemical content must be monitored for each HRSG. Since the system load for any utility is constantly changing, combined cycle power plants are required, like other power producing plants, to be dispatched, or provide load as required to the electrical grid. This means the power plant will not operate at a fixed load, but will constantly be modulating load to meet the system demand. To increase load, supplementary firing (additional fuel burned at or near the inlet to the HRSG to add energy to the exhaust gases) can be accomplished. However, this is detrimental to overall plant efficiency. This is noted by Rice in patent U.S. Pat. No. 5,628,183 with references to Westinghouse and General Electric studies. Moore in patent U.S. Pat. No. 5,649,416 states that
xe2x80x9cSupplemental firing of the heat recovery steam generator can increase total power output and the portion of the total power produced by the steam turbine, but only with a reduction in overall plant thermal efficiency.xe2x80x9d
Therefore, it is common in combined cycle plants to see little or no supplemental firing used. Therefore, to change and meet varying system loads, the GTs are brought from full load to part load operation.
As well as increasing emission levels as previously mentioned, this part load operation also has a detrimental effect on efficiency. FIG. 7 is a representative curve of GT efficiency versus load. At 100% load it consumes 100% fuel, however, at 60% load, it consumes 70.5% of full load fuel. This is an increase of 17.5% in specific fuel consumption. For large central power plants, this factor equates to significant added fuel costs. In addition, operation at part load on the GT typically increases the emission levels for the most difficult criteria pollutant, NOX. Part load operation of the GT also changes the exhaust gas flow through the HRSG. This change in flow upsets the heat transfer in the HRSG since this device is constructed with fixed heat exchange surface area. This phenomenon, as well as reduced GT efficiency, contributes to poorer overall efficiency at part load operation. If part load operation changes temperatures in the HRSG significantly, this could lead to ineffective operation of the SCR.
Steam Turbine Exhaust End Loading
Besides inlet pressure and temperature limitations, another common limitation for the steam turbine (ST) is the exhaust end loading. This essentially is a function of two parameters, exhaust end flow and exhaust pressure. These two factors essentially determine the volumetric flow through the last stage blading of the ST. For optimum operation, there is a range of volumetric flow typically specified by the ST manufacturers. As this volumetric flow increases, larger blades and/or more exhaust sections may be required.
However, due to mechanical limitations (centrifugal force), once the largest available blade volumetric limits are reached, more sections and more blades must be added to the exhaust end of the ST to accommodate this flow. This adds to the installed cost and increases the real estate requirements of the ST. Due to its configuration, a conventional combined cycle sends HP steam to the ST HP inlet, then adds steam from the IP section of the HRSG to this flow at the ST IP section inlet, then adds more steam from the LP section of the HRSG to this flow at the ST LP section inlet.
Therefore, in this arrangement, the HP and IP sections of the ST see relatively lower flows and lower volumetric efficiencies than the LP section. This arrangement leads to STs that are at or near the exhaust end limit of the ST. This provides for little in the way of temporary capacity extension for peak power production and leaves little or no ability to uprate (increase) the ST in the future to a higher power rating. Overall, this ST arrangement is less efficient than conventional steam plant STs since the HP and IP sections have low volumetric flows.
In GE informative document GER-3582E (1996), entitled xe2x80x9cSteam Turbines for STAG(trademark) Combined Cycle Power Systemsxe2x80x9d, by M. Boss, the author discusses the exhaust end loading that is associated with STs in the prior art:
xe2x80x9cExhaust sizing considerations are critical for any steam turbine, but particularly so for combined-cycle applications. There are usually no extractions from the steam turbine, since feedwater heating is generally accomplished within the HRSG. Generation of steam at multiple pressure levels (intermediate pressure and/or low-pressure admissions to the turbine downstream of the throttle) increases the mass flow as the steam expands through the turbine. Mass flow at the exhaust of a combined cycle unit in a three-pressure system can be as much as 30% greater than the throttle flow. This is in direct contrast to most units with fired boilers, where exhaust flow is about 25% to 30% less than the throttle mass flow, because of extractions from the turbine for multiple stages of feedwater heatingxe2x80x9d.
Real Estate
A combined cycle installation, although typically smaller than conventional steam plants, still occupies a large area. The HRSGs with their stacks are particularly large and require a great deal of floor area (the HRSG for one Westinghouse model 501G gas turbine is approximately 40 feet wide, 70 feet high, and 200 feet long). With the trend towards deregulation of electrical power, plant owners will be seeking the ideal site for their power plants. In many instances, this is near to the electrical load, which is usually in either an urban or industrial area. This puts the power plant close to the end user of electricity, and eliminates the need for high voltage transmission lines (which also require large amounts of real estate). However, available real estate for a large combined cycle power plant may be difficult and expense to attain in these areas.
Some prime real estate for these combined cycle power plants will be existing power plants that can be repowered as combined cycle facilities. These sites have the advantage of being properly zoned with the necessary electrical and mechanical infrastructure. The drawback is that the site may lack the necessary real estate for a combined cycle repowering project. Therefore, it is desirable from a space efficiency viewpoint as well as from a cost perspective to keep plants as small as possible.
Noise/Public Acceptance
Public acceptance is becoming increasing difficult for many utility power plant projects. Factors such as noise, traffic increase, unsightliness, pollution, hazardous waste concerns, and others contribute to public disapproval of power plants in close proximity to populated areas. A plant that can be built smaller, quieter, with less equipment, lower emissions, and maintain a low profile is preferred over a larger, more obvious plant. Therefore, more compact, higher xe2x80x9cpower densityxe2x80x9d (power per unit volume) combined cycle power plants are desired.
However, to meet the current trends in demand for power consumption, conventional power plants being constructed today simply replicate existing proven plant designs to meet the increased energy consumption demand. No attention is currently being given to the issue of whether plants may be redesigned to consider the ancillary issues associated with the public acceptance of the plants themselves.
Heat Rejection
Both conventional steam and combined cycle power plants require some form of heat rejection. This is typically to condense the low-pressure steam from the ST exhaust back into water. This heat rejection can be to the air, river, lake, or other xe2x80x9creservoirxe2x80x9d that will absorb the heat. Since this heat rejection will have an effect on the local environment and possibly the local biological life (i.e., fish in a river), methods to reduce heat rejection requirements are always in demand.
Gas Turbine Performance Decay
Although combined cycle power plants demonstrate high efficiencies, these efficiencies are for xe2x80x9cnewxe2x80x9d power plants. Since the combined cycles in the prior art are primarily GT based, their efficiency levels are very susceptible to GT performance decay, a phenomenon in which the efficiency of the GT degrades substantially (2% to 6%) within only a year or two of operation. This can be a significant factor in the cost of fuel as the overall combined cycle efficiency also degrades as the GT performance decays.
Accordingly, the objects of the present invention are to circumvent the deficiencies in the prior art and affect the following objectives:
1. Provide a combined cycle power plant that has more design flexibility than current offerings so that developers can have state-of-the-art facilities, but purchase them at the capacity they need.
2. Reduce overall fuel consumption at rated output, but especially at part load conditions, as the plant will likely spend only a small fraction of its operating time at rated load.
3. Reduce installed cost of the power plant such that the debt service is substantially reduced and that financing by a bank or other lending institution is much easier for the owner.
4. Leverage the time value of money with regards to capital, maintenance, and fuel costs to make the creation of power plants more economically efficient and hopefully reduce the overall cost of electric power generation.
5. Provide the ability for the power plant to meet peak demand loads without sacrificing normal operation efficiency or significantly increasing the installed cost.
6. Reduce inefficiencies and losses associated with the transmission of power over long distances.
7. Increase the overall reliability of the electrical grid by permitting electrical power to be generated local to the demand during times of peak demand loads.
8. Reduce OandM costs. Besides fuel costs, the objective is also to reduce costs for maintenance, supplies, inventory, insurance, and other operating expenses.
9. Reduce the need for fuel gas compression.
10. Improve reliability.
11. Reduce air consumption and air filtering requirements.
12. Lower emissions of criteria pollutants, especially NOX.
13. Minimize the discharge of water from HRSG blowdown and other sources.
14. Utilize controls to the maximum extent feasible to increase efficiency, reliability, and heat recovery.
15. Simplify operation and devise methods and/or strategies to increase part load efficiency and reduce emission levels.
16. Optimize the ST efficiency by utilizing designs with improved volumetric efficiency and excess capacity to meet peak power demands.
17. Conserve space and land mass required to house the power plant by designing a compact, high power density arrangement.
18. Reduce noise, size, space requirements, and equipment to minimize the effect the power plant has on local residents and the community.
19. Keep heat rejection to a minimum.
20. Provide for economic and space efficient retrofit of existing steam power plant and combined cycle installations so as to reduce capital costs and the economic burden associated with major equipment additions and added real estate requirements.
21. Provide economic incentive for new plant construction to use environmentally friendly designs.
22. Design combined cycle power plants that are less susceptible to gas turbine performance decay.
These objectives are achieved by the disclosed invention that is discussed in the following sections.
Briefly, the invention is a system and method permitting the use of fewer and/or smaller gas turbines (GTs) and heat recovery steam generators (HRSGs) in a combined cycle application. This conventional combined cycle equipment is replaced by a larger steam turbine and continuously fired heat recovery steam generators to provide a variety of economic, energy conservation, and environmental benefits.
Present technology utilizes multi-pressure HRSGs to maximize the heat recovery from exhaust gases of a GT. This arrangement is commonly used because the prior art teaches away from using continuously fired HRSGs because of the common belief that these configurations have lower thermal efficiencies. Despite this commonly held belief, the present invention teaches that continuously fired HRSGs can be configured with thermal efficiencies on par or better than current combined cycle practice. However, to obtain this level of efficiency, the continuously fired HRSGs and ST must be configured and designed differently than current practice.
In several preferred embodiments of the present invention, the GTs are unchanged from the present art and exhaust to an HRSG. This HRSG, however, is designed as a single pressure level steam generator (SPLSG) (or primarily a single pressure level) which is optimized for continuous firing to produce higher pressure steam than in conventional combined cycle practice. In addition, the HRSG is designed to have controlled feedwater flows through the economizer/feedwater sections to maximize heat recovery. Also, the ST is designed as a larger unit, typical of that which would be found in a conventional Rankine Cycle plant, with reheat and conventional ST extraction steam fed feedwater heaters to maximize plant thermal efficiency. This benefit of a larger ST typical of a conventional steam plant is described by Moore in patent U.S. Pat. No. 5,649,416 which is assigned to General Electric:
xe2x80x9cConventional steam power plants benefit in both lower cost and higher efficiency through the economies of scale of large ratings. A traditional rule of thumb regarding cost is that the doubling of plant rating results in a ten percent reduction in cost. The cost of one large generating unit according to this rule would be expected to cost on the order of ten percent less than that for a plant with two half-size units. Efficiency is also improved with increased size and power ratings. As with all turbomachinery, the internal efficiency of the steam turbine is a strong function of the inlet volumetric flow, which is directly proportional to the rating. Also, as is well known, the thermal efficiency of the Rankine Cycle increases with the pressure at which steam is generated. Increasing pressure, however, reduces the volumetric flow of the steam at the turbine inlet, reducing the internal expansion efficiency. The offsetting effect in overall efficiency, however, is much greater at low volumetric flow than at high volumetric flow. Therefore, an additional performance related benefit of increasing turbine size is that higher steam throttle pressure can be utilized more effectively.xe2x80x9d
With the use of ample supplemental firing in the HRSG, the bottoming cycle with the present invention is given the liberty to be more independent from the GT operation. Therefore, the GTs can be operated at full load while the overall plant load is modulated over a wide range of its full load capability by only changing the supplemental firing rate and the STs load. This increases the overall plant rating when utilizing a given set of GTs, provides flexibility for the combined cycle plant rating through variation in the rate of supplemental firing, as well as increases the overall plant thermal efficiency at part load. In addition, it simplifies operation, and has the potential to reduce emissions.
By designing the HRSGs to be capable of firing to 2400xc2x0 F., an exemplary single 2-on-1 arrangement of two GTs and one large ST replaces two 2-on-1 arrangements (4-on-1 arrangements are typically not available when reheat is utilized due to balancing problems on the reheat lines). This exemplary configuration saves two GTs, two HRSGs, one ST, three switchgear, three transformers, and the accessories, real estate, and maintenance required to support this equipment. Capital costs for the power plant in US$/kW are thus greatly reduced using the teachings of the present invention.
All this is accomplished by utilizing proven turbomachinery technology and hardware. The continuously fired HRSG with a single pressure is a novel concept for this application, but is not beyond technological practice nor capability for implementation in the current art. Therefore, there are little or no compromises in reliability. The general architecture for several preferred embodiments of the present invention is illustrated in FIG. 13, with several exemplary embodiments having more detail illustrated in FIG. 9 and FIG. 15.
Improvements Over the Prior Art
The present invention solves the problems present in the prior art by achieving the following objectives:
1. Providing more design flexibility in the combined cycle power plant so that developers can still achieve state-of-the-art efficiency, but yet specify the capacity they need.
2. Reducing overall fuel consumption by improving both full load and part load efficiency.
3. Reducing installed costs by increasing the power density of the installation (more power output per a given amount of equipment).
4. Reducing the overall cost of producing electricity by reducing the three major factors associated with its production: fuel consumption, capital costs, and maintenance costs.
5. Provide temporary capacity for attaining peak loads by utilizing supplemental firing to produce more steam, as well as having the option to operate the ST at overpressure (inlet pressure slightly above rated) and reducing extraction steam flow to the feedwater heaters.
6. Increasing the efficiency of the power grid by permitting local generation of power during periods of peak loading. By permitting local power generation during these peak periods, inefficiencies associated with xe2x80x9cimportingxe2x80x9d power from other areas of a given country (and outside a country) are reduced or eliminated. (These are energy losses associated with transmitting power through power transmission lines).
7. Increasing the reliability of the electrical power grid by reducing the long haul transmission of electrical power during times of peak power loading.
8. Reducing OandM costs, primarily by reducing the amount of equipment and systems and utilizing equipment that has lower maintenance costs per kWh produced (low maintenance cost STs versus high maintenance cost GTs).
9. Minimizing the need for fuel gas compression by utilizing fewer GTs and GTs with lower fuel gas pressure requirements in the cycle in conjunction with a larger ST.
10. Improving reliability by reducing the complexity of the power plant design.
11. Reducing air consumption by utilizing fewer GTs.
12. Lowering emissions of criteria pollutants, especially NOX, by operating the GTs at a steady, low emissions operating point, utilizing cleaner GTs, and utilizing fewer GTs.
13. Minimizing blowdown and other discharge through higher efficiency cycles that require less steam flow per kW of electricity generated.
14. Utilizing controls to increase efficiency, reliability, and heat recovery.
15. Simplifying operation by running the GTs at full load over a wide range of operation (total combined cycle plant output) and reducing HRSG pressure levels to only one.
16. Maximizing ST efficiency by increasing volumetric flows, especially in the HP and IP sections.
17. Conserving space and land mass with less equipment and higher power density designs.
18. Reducing noise, size, and space requirements with less equipment.
19. Keeping heat rejection to a minimum by utilizing high efficiency cycles with less heat rejection per kWh produced.
20. Providing a combined cycle design that is more compatible with existing steam power plants allowing for more compact and cost effective retrofits of these existing plants to high efficiency combined cycle technology.
21. Minimizing air consumption, emissions of criteria pollutants, and heat rejection to the atmosphere, but providing these environmental benefits with lower cost than the conventional combined cycles.
22. Reducing the impact of gas turbine performance decay by utilizing a combined cycle power plant that is less dependent upon the gas turbines and their efficiency.